Infection-based self-configuration in agent societies

ABSTRACT

Norms have become a common mechanism to regulate the behavior of agents in multi-agent systems (MAS). However, establishing a stable set of norms is not trivial, particularly in dynamic environments, under changing (and unpredictable) conditions. We propose a computational model that facilitates agents in a MAS to collaboratively evolve their norms, reconfigure themselves, to adapt to changing conditions. Our approach borrows from the social contagion phenomenon to exploit the notion of positive infection: agents with good behaviors become infectious to spread their norms in the agent society. By combining infection and innovation, a mechanism allowing agents exploring new norms, our computational model helps MAS to continuously stabilize despite perturbations.

REFERENCES

• 1: Sherief Abdallah , Victor Lesser, Multiagent reinforcement learning and self-organization in a network of agents, Proceedings of the 6th international joint conference on Autonomous agents and multiagent systems, May 14-18, 2007, Honolulu, Hawaii [doi>10.1145/1329125.1329172]
• 2: R. Axelrod. An evolutionary approach to norms. The American Political Science Review, 80(4):1095--1111, 1986.
• 3: Thomas Bäck, Evolutionary algorithms in theory and practice: evolution strategies, evolutionary programming, genetic algorithms, Oxford University Press, Oxford, 1996
• 4: R. Burt. Social contagion and innovation: Cohesion versus structural equivalence. American J. of Sociology, 92:1287--1335, 1987.
• 5: R. Conte and M. Paolucci. Intelligent social learning. Artificial Society and Social Simulation, 4(1):1--23, 2001.
• 6: Il-Kwon Jeong , Ju-Jang Lee, Evolving multi-agents using a self-organizing genetic algorithm, Applied Mathematics and Computation, v.88 n.2-3, p.293-303, Dec. 30, 1997 [doi>10.1016/S0096-3003(96)00337-2]
• 7: Zbigniew Michalewicz, Genetic algorithms + data structures = evolution programs (3rd ed.), Springer-Verlag, London, 1996
• 8: D. Moriarty, A. Schultz, and J. Grefenstette. Evolutionary algorithms for reinforcement learning. Artificial Intelligence Research, 11(1-1):241--276, 1999.
• 9: J. Pujol, J. Delgado, R. Sangüesa, and A. Flache. The role of clustering on the emergence of efficient social conventions. In IJCAI 2005, pages 965--970, 2005.
• 10: S. Sen and S. Airiau. Emergence of norms through social learning. In IJCAI 2007, pages 1507--1512, 2007.
• 11: Yoav Shoham , Moshe Tennenholtz, On the emergence of social conventions: modeling, analysis, and simulations, Artificial Intelligence, v.94 n.1-2, p.139-166, July 1997 [doi>10.1016/S0004-3702(97)00028-3]
• 12: A. Walker and M. Wooldridge. Understanding the emergence of conventions in multi- agent systems. In ICMAS 1995, pages 384--389, 1995.

INDEX TERMS

Primary Classification:
  I. Computing Methodologies
  I.2 ARTIFICIAL INTELLIGENCE
        I.2.11 Distributed Artificial Intelligence
            Subjects: Multiagent systems

General Terms:
Algorithms, Experimentation

Keywords:
distributed evolutionary algorithm, infection-based algorithm, norms' evolution